The Genomics of Hybrid Speciation

The Genomics of Hybrid Speciation

Cassandra Nicole Trier

Dissertation presented for the degree of Philosophiae Doctor (PhD)

2018

Centre for Ecological and Evolutionary Synthesis Department of Biosciences

Faculty of Mathematics and Natural Sciences

© Cassandra Nicole Trier, 2018

Series of dissertations submitted to the

Faculty of Mathematics and Natural Sciences, University of Oslo No. 2020

ISSN 1501-7710

All rights reserved. No part of this publication may be

reproduced or transmitted, in any form or by any means, without permission.

Cover: Hanne Baadsgaard Utigard.

Print production: Reprosentralen, University of Oslo.

Science is magic that works.

-Kurt Vonnegut

Acknowledgements

This PhD has been a crazy journey and the truth is, I would have never made it through without support from so many wonderful people along the way. First of all, I would like to thank my supervisors Glenn- Peter Sætre and Kjetill S. Jakobsen. Glenn, you didn’t know what you were signing up for when I knocked on your door as a new masters student looking for a research group! Now, eight years later, I can say that it has been a pleasure being in your research group and I will always be grateful for the guidance and support you have given me through the years. Kjetill, you have always been there if I need advice or just someone to talk to about my project. Your up-beat personality and great sense of humor make you easy to talk to and I always enjoy our chats. Thank you so much for your support.

I would also like to thank my co-authors and friends in the

sparrow group for so many great discussions that left me inspired in my own work. In particular, thank you Angelica, Anna R., Camilla, Caroline, Fabrice, Jo, Mark and Melissah for being a constant source of encouragement through the years. You have always made me feel like my opinion is valuable even when I have doubted so myself; it has meant more to me than you know.

The Centre for Ecological and Evolutionary Synthesis (CEES) has

been such a fantastic work environment where I have made life-long

friends. I feel so fortunate to have worked with such an amazing group

of people. Thank you to all my co-workers for the ridiculous lunchtime

conversations and making CEES a fun place to work. A special thanks

to Anders, Anna M., Bastiaan, Boris, Eric, Heidi, Helle, Inger Maren,

Katie, Kjetil, Luis, Ole Kristian, Olja, Pernille, Ryan, Sanne and Unni for

the fun that also extended outside of work at barbecues, parties,

hikes, kids play-dates, you name it.

To my knitting girls, your love and constant support has meant the world to me. Life is so much easier with friends like you. Thank you for keeping me sane and bringing so much joy and laughter into my life.

I have never stopped missing my friends from California. To Alex, Anna H., Jack, Jiro, Kevin, Nicole, Matt and Tarek, thank you for being life-long friends that keep me grounded and have put up with me all these years!

I would also like to thank all my extended family that held back from teasing me too much about “taking care of the birds” all this time.

Thank you for your continuous love and support. A special thank you to Grandpa Tom who always believed in me and I know would be proud.

To Mom and Dad, it has only been with your unconditional love that I had the courage to come to Norway in the !rst place. You

constantly sacri!ce so I can have a good life, even when it means your only daughter moves across the world. Thank you for always being there for me, believing in me and pushing me to follow my passion.

To my dear husband, Tore, I’m not sure any words su"ce. Not only have you stuck by me as my number one fan throughout the PhD, but you have been my o"cemate, co-author and best friend. People wonder how I could share an o"ce with my husband and the truth is, it was easy because you and our sons are always the highlight of my day.

Thank you for quite simply everything along the way, I couldn’t have done it without you.

And last but not least, my sons. Oliver and Eirik, thank you for everyday reminding me of what is really important in life.

Oslo, June 2018

Table of Contents

List of Papers 1

Summary 2

Introduction 3

Background 3

Hybrid speciation 4

Detecting hybrid origin 6

Hybrid zones 8

The genomic architecture of hybrid species 11

The Passer sparrow system 13

Aims 15

Paper Summaries 16

Discussion 19

Patterns of admixture in a hybrid species 19

Reproductive isolation from the parents 21

Selection within the hybrid lineage 23

Future perspectives and concluding remarks 23

Acknowledgements 25

References 25

Paper I

Paper II

Paper III

List of Papers

Paper I.

C. N. Trier*, J. S. Hermansen*, G.-P. Sætre, R. I. Bailey. (2014) Evidence for mito-nuclear and sex-linked reproductive barriers between the hybrid Italian sparrow and its parent species. PLoS Genetics

10:e1004075.

Paper II.

T.O. Elgvin*, C. N. Trier*, O.K. Tørresen, I. Hagen, S. Lien, M. Ravinet, H.

Jensen, G.-P. Sætre. (2017) The genetic mosaicism of hybrid speciation. Science Advances 3:e1602996.

*These authors contributed equally to the paper.

Paper III.

A. Runemark, C. N. Trier, F. Eroukhmano!, J.S. Hermansen, M.

Matschiner, M. Ravinet, T.O. Elgvin, G.-P. Sætre. (2018) Variation and

constraints in hybrid genome formation. Nature Ecology and Evolution

2:549-556.

Summary

The central question of this thesis is how hybridization can lead to the formation of new species. To approach this question, I used the hybrid Italian sparrow (Passer italiae) to explore the consequences of hybridization on genomic architecture and reproductive isolation at the genomic level. With the use of transcriptomic data and cline analyses in Paper I, we investigated if there is evidence for reproductive barriers between the hybrid species and its parents, the house sparrow (P. domesticus) and Spanish sparrow (P.

hispaniolensis); and if so, which genes or genomic regions may be involved?

We found that there is evidence of reproductive isolation between the Italian sparrow and its parent species and that Z-linked genes and mito-nuclear gene complexes play an integral role.

In Paper II, we sought to examine the genomic architecture of the Italian sparrow in comparison to its parents. This "rst consisted of de novo assembling a high-quality reference genome of the house sparrow. We then mapped whole-genome sequencing data from populations of the parent house and Spanish sparrows, as well as the Italian sparrow to the reference genome. By using comparative genomics, we were able to characterize patterns of admixture and di!erentiation in the Italian sparrow genome in relation to both parent species. We show that the genomic landscape of the Italian sparrow is highly heterogenous with regions inherited alternately from either parent across the genome in a mosaic pattern. High divergence regions between the Italian sparrow and either of its parents were found to be disproportionately located on the Z chromosome and genes involved in body patterning, beak morphology and the immune system were over- represented in these regions. We also found regions where the Italian sparrow is divergent from both parents that may represent areas of novel divergence in this homoploid hybrid lineage.

In Paper III, we utilized genomic data from multiple geographically

isolated Italian sparrow populations from di!erent Mediterranean islands to

explore the extent to which the hybrid genome can vary. We "nd that there is

variation in the genomic combinations that compose a functional hybrid species, yet there are some areas that are invariably inherited from one parent. These regions of genomic constraint are over-represented on the Z chromosome and hold candidate incompatibility loci involved in DNA repair and mito-nuclear function.

Overall, this dissertation helps demonstrate how a new species can arise via hybridization by painting a picture of how admixture can shape the genomes of di!erentiated populations and lead to the formation of reproductive barriers.

Introduction

Background

Hybridization is widespread in nature and can have various impacts on species diversi"cation. Traditionally, interbreeding between distinct populations has been viewed as a detriment to species divergence and a

‘biological mistake’ (Mayr 1932; Fisher 1930). Yet, around 10% of animal and 25% of plant species are known to hybridize (Mallet 2007), making hybridization a prominent feature in nature. Hybridization has even been shown to have shaped the genome of our own species (Sankararaman et al.

2014). In recent years, it has become apparent that hybridization can also serve as a source of genetic variation promoting diversi"cation (Abbott et al.

2013; Seehausen et al. 2014). Since new genetic variation can have adaptive

potential, species divergence can alternatively be facilitated through

hybridization rather than impeded (Grant and Grant 1994; Heliconius

Genome Consortium 2012). In fact, hybridization is particularly common

among rapidly radiating groups (Mallet 2007; Grant et al. 2015) as it can allow

for rapid local adaption through the introgression of new variation subject to

selection (Lamichhaney et al. 2018). In some instances, hybridization can

even give rise to new species through the recombination of parental

genomes, leading to a third population of mixed ancestry that remains distinct from both its parents (Rieseberg 1997; Mallet 2007; Mavarez and Linares 2008). Thus, the creative role of hybridization in evolution spans a continuum from adaptive introgression to hybrid speciation. This thesis focuses on the latter and aims at gaining a better understanding of how hybridization can play a primary role in the origin of new species.

Hybrid speciation

When two di!erentiated populations mate, hybrid o!spring possessing novel, mixed genotypes can arise (Rieseberg 1997; Buerkle et al. 2000;

Coyne and Orr 2004). These hybrid o!spring may have reduced "tness relative to their parent taxa (Mayr 1963; Coyne and Orr 2004) due to genetic incompatibilities leading to inviability or infertility in "rst generation (F1) hybrids (Arnold and Hodges 1995) or ecological intermediacy (Schluter 1993; 1995; Coyne and Orr 2004). However, empirical studies have shown that hybrid genotypes possess a wide range of "tnesses and that hybrids can have equivalent or higher "tness than their parents (Arnold and Hodges 1995; Arnegard et al. 2014).

Hybridization provides the means in which new genetic variation from multiple loci with adaptive potential are transferred simultaneously.

Therefore, adaptive evolution may proceed more rapidly following

hybridization than would be expected from mutations alone (Grant and Grant

1994; Heliconius Genome Consortium 2012; Abbott et al. 2016). F1 hybrids

can also experience an increase in growth rate, size and reproductive

success, known as heterosis or hybrid vigor (Arnold and Hodges 1995),

which may constitute an adaptive advantage relative to its parents. Though

the genetic basis for hybrid vigor is subject to much debate, recent

hypotheses suggest it may be attributed to complementary interactions of

alleles at multiple loci (epistasis) (Baack and Rieseberg 2007). Hybrid vigor is

often broken down in subsequent generations as recessive alleles become

exposed and parental gene combinations are broken up via recombination (Dobzhansky 1948; Templeton 1981; Felsenstein 1981). Yet, if the hybrid lineage is able to persist and develop reproductive barriers against both its parents, it has the potential to become its own species. This process in which interspeci"c hybridization gives rise to novel species is known as hybrid speciation.

An important aspect of hybrid speciation is the hybrid karyotype in relation to its parents. Allopolyploid hybrids maintain a di!erent number of chromosome complements than their parents and consequently develop immediate reproductive isolation. This, combined with potential heterosis, may lead to the establishment of new lineages and has frequently been observed in plant taxa (Rieseberg 1997; Mallet 2007; Rieseberg and Willis 2007; Hegarty and Hiscock 2008; Soltis and Soltis 2009). In contrast, homoploid hybrids, which share the same ploidy level as their parents, may struggle to develop reproductive barriers strong enough to remain distinct from their parents (Baack & Rieseberg 2007). For this reason, homoploid hybrid speciation has historically been considered to be a rare outcome of hybridization (Mallet 2007; Schumer et al. 2014), particularly in animals (Mavarez and Linares 2008). The hybrid must "rst escape genetic incompatibilities and "tness loss, and then the homogenizing e!ect of gene

#ow from its parents; despite complementary ploidy levels and the fact that their parent’s reproductive barriers were su$ciently weak that they interbred in the "rst place (Buerkle et al. 2000; Coyne and Orr 2004). In some instances however, the recombination of parental alleles from initial hybridization could trigger the formation of reproductive barriers between the parents via ecological divergence (Gross and Rieseberg 2005), assortative mating (Mavarez et al. 2006; Melo et al. 2009) or genetic incompatibilities (Rieseberg 1997; Schumer et al. 2014; Abbott et al. 2016). In particular, one mechanism predicted to allow for homoploid hybrid speciation is transgressive segregation - the production of hybrid traits outside the range of its parents - that enable the hybrid to colonize new ecological niches (Rieseberg et al.

1999; Mallet 2007). While there are few well-documented cases of

homoploid hybrid animal species, new methods have led to a growing

number of empirical examples in the past decade in #ies (Schwarz et al.

2005), bats (Larsen et al. 2010), butter#ies (Mavarez and Gonzalez 2006;

Kunte et al. 2011; Heliconius Genome Consortium 2012), birds (Elgvin et al.

2011; Hermansen et al. 2011) and "shes (Salzburger et al. 2002; Meyer et al.

2006; Keller et al. 2013) suggesting that homoploid hybrid speciation may be more common than initially thought (Mavarez and Linares 2008; Abbott et al.

2013).

Detecting hybrid origin

For many years, con#icting phylogenetic trees from nuclear and organellar markers have been used to test for hybrid ancestry in proposed hybrids (Bullini 1994; Dowling and Secor 1997; Soltis and Soltis 2009). In the past, this has proven problematic when few markers are used because incomplete lineage sorting can also produce discordant phylogenies (reviewed in Ballard and Rand 2005). Genetic mosaicism, where there is evidence of alternating inheritance from two parental lineages, has been considered strong evidence for hybrid speciation, but it is also di$cult to demonstrate with few markers since hybridization upon secondary contact can give the same signal (vonHoldt et al. 2011; Schumer et al. 2014). As data sets have gotten bigger and genome-wide markers or whole genome sequencing are becoming commonplace, this has allowed for more robustly testing of hybrid ancestry hypotheses. However, it has also raised new questions in regards to what criteria are needed to determine if a taxon is derived from hybrid speciation.

As the scale of genetic data has gotten larger, it has become

increasingly apparent that hybridization is a prominent feature in nature

(Mallet 2007; Abbott et al. 2013) and the number of cases of proposed

hybrid species has increased dramatically (Schumer et al. 2013; 2014). This

has led to some confusion in the literature as to what constitutes ‘hybrid

speciation’ (Mallet 2007; Mavarez and Linares 2008; Jiggins et al. 2008;

Abbott et al. 2013; 2016). Schumer et al. 2014, argued that many purported examples of hybrid speciation do not have strong enough evidence and proposed criteria required in order to conclude that a species is product of hybrid speciation. They suggest there needs to be evidence of i) reproductive isolation between the hybrid lineage from its parents, ii) signatures of hybridization in the genome, and iii) reproductive isolation being a direct product of hybridization. Since speciation is a continuum and de"ning the point at which two divergent populations become a species has proven di$cult (Mallet 1995), it should come as no surprise that there is a debate on what evidence is required for a species to be considered of hybrid origin. In particular, Feliner et al. 2017 have criticized Schumer et al. 2014, speci"cally criterion iii, arguing that if a hybridization led to an established, ecologically and morphologically distinct hybrid lineage, this should be considered hybrid speciation regardless of whether or not hybridization directly led to reproductive isolation.

There does however appear to be some agreement that support for hybrid speciation should include evidence of both hybrid ancestry and reproductive isolation from the parental lineages. Genomic data provides great opportunities to investigate these aspects of proposed hybrid species.

For instance, since many more genomic regions can be sampled, the

chances of identifying the areas involved in reproductive isolation are higher

(Schumer et al. 2014). Also, large-scale signatures of hybridization such as

genomic mosaicism can help distinguish between patterns of hybridization

from those driven by incomplete lineage sorting, genetic drift or selection

(Nice et al. 2013; Schumer et al. 2014). While there is certainly a gray area in

what is considered hybrid speciation, what is perhaps more fruitful than

squabbling over its de"nition is using newly available genomic data to better

understand the processes involved in hybridization leading to the formation

of new species. In this thesis, the goal was to do just that; use genomic data

to explore how hybridization can result in a new, distinct and reproductively

isolated lineage.

Hybrid zones

Hybrid zones are regions where “genetically distinct groups of individuals meet and mate, resulting in at least some o!spring of mixed ancestry” (Harrison 1993). They provide an excellent opportunity to study reproductive barriers as they are in e!ect natural laboratories where parental ancestry blocks are broken down and new combinations of genes are exposed to selection (Barton and Hewitt 1989; Buerkle et al. 2000; Payseur and Nachman 2005; Harrison and Larson 2014).

Postzygotic reproductive barriers between a hybrid and its parents are often the result of Bateson-Dobzhansky-Muller (BDM) incompatibilities (Bateson 1909, Dobzhansky 1936; Muller 1940; Orr 1996; Turelli and Orr 1995). Under a BDM model, hybrids may receive alleles from two populations that have diverged at di!erent loci without su!ering a loss in "tness in either population, but when these divergent alleles introgress into a new genomic background, they may interact poorly resulting in a "tness loss (Figure 1).

This could be due to selection against heterozygotes, selection against

certain alleles in a foreign genetic background or a combination of the two

(Barton 2001; Buerkle and Rieseberg 2001; Baack and Rieseberg 2007). The

rate of gene #ow of an allelic variant in a hybrid zone is consequently

determined by its e!ect on "tness and linkage disequilibrium to other

genomic regions (Gompert and Buerkle 2011b). This is true for all loci

experiencing divergent selection in the hybrid zone and is expected to be

especially strong for BDM incompatibilities as they often have large "tness

e!ects. Therefore, hybrid zone analyses provide an opportunity for studying

genomic regions that may drive speciation.

Figure 1. BDM model of genetic incompatibilities. After populations of the ancestral genotype split, new mutations arise and reach !xation without a

!tness loss. When the divergent populations meet again and mate, new combinations of incompatible alleles result in a !tness loss in the hybrid o"spring.

By examining patterns of di!erentiation of loci across hybrid zones, it is possible to identify loci that exhibit reduced introgression and consequently are candidates for being involved in reproductive barriers between species (Barton and Hewitt 1985; Gompert and Buerkle 2011a; Gompert et al. 2012).

One manner of quantifying di!erential rates of introgression among loci in hybrid zones is through the use of cline theory. Cline theory measures the gradient in a trait or allele frequency across a geographic or genomic range.

In terms of gene #ow, cline theory can be used to measure rates of gene #ow across a geographic range (geographic clines) (Szymura and Barton 1986;

Carling and Brum"eld 2008; Teeter et al. 2010; Taylor et al. 2012) or into di!erent genomic backgrounds (genomic clines) (Szymura and Barton 1986;

Gompert and Buerkle 2011b; Taylor et al. 2014). Cline theory predicts that the width of a cline is dependent on a balance between selection and dispersal (Slatkin 1973; Barton and Hewitt 1985). Neutral alleles are predicted to introgress proportionally to dispersal distance, i.e. the gene #ow out of the hybrid zone. Meanwhile, alleles that reduce "tness or contribute to assortative mating are expected to introgress less due to strong selection creating narrow clines (Figure 2) (Barton and Hewitt 1989; Harrison 1993;

Buerkle and Lexer 2008; Gompert and Buerkle 2011b). If partial reproductive isolation exists between two taxa at a given locus, steep clines in allele frequencies indicative of reduced introgression would be expected in hybrid zones (Slatkin 1973; Nagylaki 1975; May et al. 2015).

A B A B

A b

A B a B A B

A b

A b a B a B

X

a B

Ancestral genotype New mutations Fixation of mutations Hybrid offspring

X

In the context of hybrid speciation, advantageous alleles are expected to spread rapidly and be driven to "xation in the hybrid lineage while BDM incompatibilities are expected to be purged via recombination (Buerkle and Rieseberg 2008). Once the hybrid lineage diverges from one or both of the parents in areas of the genome, novel incompatibility factors may then arise if the hybrid backcrosses with its parent taxa (Buerkle and Rieseberg 2008).

Only a few generations of recombination may be su$cient in creating a high

"tness hybrid lineage with reproductive isolation from its parents (Buerkle et al. 2000; Buerkle and Rieseberg 2008). Detection of loci experiencing reduced introgression between a hybrid and its parental lineages in hybrid zones can therefore provide insight into how reproductive barriers form in the hybrid speciation process.

Figure 2. Example of geographic clines. Two populations !xed for di"erent alleles hybridize creating steep (red) and shallow (blue) clines in allele frequencies depending on the strength of selection in relation to dispersal.

Allele frequency

Distance Allele A Hybrid Zone Selection against

Selection against dispersal

A

a

Allele a

The genomic architecture of hybrid speciation

Genomic analyses have consistently shown that there is variation in the permeability of foreign alleles across the genomic landscape (Payseur and Nachman 2005; Baack and Rieseberg 2007; Harrison and Larson 2014;

Payseur and Rieseberg 2016). The term ‘genomic islands’ has become commonly used to refer to regions where gene #ow is restricted and thus divergence between lineages is relatively high (Turner et al. 2005). Genomic islands are suggested to develop through selective sweeps of favorable variants along with physically linked surrounding neutral variation (Via and West 2008) while the homogenizing e!ect of gene #ow reduces di!erentiation in the rest of the genome (Burri et al. 2015). The sizes of these regions is expected to be in#uenced by selection and the rate of recombination (Nachman and Payseur 2012; Samuk et al. 2017; Ravinet et al.

2017) and they vary largely between hybridizing species pairs. For example, some hybridizing species have only small regions of divergence (Good et al.

2015; Toews et al. 2016) while others have substantial areas of genome-wide divergence (Ellegren et al. 2012; Parchman 2013). Identifying and characterizing high divergence areas, or ‘islands’ where introgression is reduced can therefore aid in understanding which genome regions may be involved in the maintenance of species barriers.

In particular, sex chromosomes have been considered ‘hot spots’ of species divergence as numerous studies have found higher di!erentiation and reduced introgression on sex-linked loci (Macholán et al. 2007; Teeter et al. 2010; Ellegren et al. 2012; Carneiro et al. 2013; Taylor et al. 2014). This pattern is likely attributed to i) a smaller e!ective population size, ii) lower recombination rates, iii) exposure of recessive alleles to selection in the heterogametic sex, and iv) non-random accumulation of genes involved in sex and reproduction (Charlesworth et al. 1987; Qvarnström and Bailey 2008;

Mank et al. 2010). For these reasons, sex chromosomes are often viewed as

farther along in the speciation continuum than the rest of the genome. An

example in hybridizing birds are collared (Ficedula albicollis) and pied

(Ficedula hypoleuca) #ycatchers. They exhibit higher species di!erentiation on sex chromosomes than autosomes (Ellegren et al. 2012) and evidence has shown that both male traits under sexual selection and female preference are sex-linked, thus accelerating the speciation process on sex chromosomes (Sæther et al. 2007).

As there are few well-documented cases of homoploid hybrid species, little is known about the genomics of this form of speciation including the nature of reproductive barriers and how genomic mosaicism evolves. One proposed mechanism for driving hybrid incompatibilities and species divergence is through mito-nuclear incompatibilities (Bar-Yaacov et al. 2015;

Hill 2017). The mitochondrial genome is maternally inherited and responsible for energy production via oxidative phosphorylation in animals. It is also non- recombining with a mutation rate an order of magnitude higher than the nuclear genome, yet both nuclear and mitochondrial genomes are responsible for mitochondrial function (Rand et al. 2004). Mito-nuclear dysfunction is expected to have strong e!ects on hybrid "tness as it has been shown to a!ect (among other things) aging and fertility (Camus et al.

2015; Patel et al. 2016). Therefore, strongly selected upon co-adapted mito- nuclear gene complexes are strong candidates for BDM incompatibilities and may act as potent reproductive barriers during hybridization and speciation (Hill 2017).

With whole genome sequencing becoming increasingly more

a!ordable, new opportunities have arisen to study hybrid speciation through

patterns of introgression on a genomic scale. The examination of

di!erentiation and admixture between populations across the genome can

shed light on the evolutionary processes involved in hybrid speciation

(Payseur and Rieseberg 2016). In this thesis, I use genomic data on a

proposed hybrid species to examine its genetic composition in relation to its

parents and help elucidate how a new species was formed as a product of

hybridization.

The Passer sparrow system

The Italian sparrow represents one of the few examples of vertebrate species shown to be of hybrid origin (Elgvin et al. 2011; Hermansen et al.

2011). Found on the Italian peninsula and some Mediterranean islands (Figure 3), the Italian sparrow was for many years suggested to have arisen through hybridization between the Spanish sparrow and the house sparrow (Summers-Smith 1988; Töpfer 2006) based on its intermediate male plumage coloration (Figure 3). More recently, genetic studies have provided support for this hypothesis by demonstrating that the Italian sparrow is genetically mosaic with genes on the Z chromosome nearly "xed for alternate parent’s alleles as well as mitochondrial DNA (mtDNA) nearly "xed for house sparrow inheritance (Elgvin et al. 2011; Hermansen et al. 2011).

However, these studies were preformed with small sets of genetic markers with limited power in detecting hybrid ancestry.

One proposed scenario for the origin of the Italian sparrow is that as the human commensal house sparrow’s range expanded throughout Europe

<10,000 years ago alongside the spread of agriculture during the Neolithic revolution, it encountered and hybridized with the Spanish sparrow which is believed to have already been present in the Mediterranean (Hermansen et al. 2011; Sætre et al. 2012; Ravinet et al. 2018). A challenge in studying hybrid species systems can be the lack of geographic overlap between the species, which makes it more di$cult to investigate reproductive barriers.

Yet, the Italian sparrow’s distribution overlaps with those of its parents as it

encounters and occasionally hybridizes with the house sparrow in narrow

contact zones in the Alps (Hermansen et al. 2011) and lives in sympatry with

the Spanish sparrow in a small, recently established contact zone on the

Gargano peninsula of southeast Italy (Figure 3). Furthermore,

morphologically divergent Italian sparrow populations are also present on

the islands of Corsica, Crete, Sicily and Malta (Figure 3). This unique sparrow

system, combined with modern genomic tools, therefore provides an

excellent opportunity to study how reproductive barriers have evolved in a

hybrid species and this is where my PhD research begins.

Figure 3. Distribution map and male plumage. Top: Illustrations of male plumage patterns in house (left), Italian (center) and Spanish (right) sparrows modi!ed from Svensson et al. 1999. Bottom: Distribution map of house, Italian and Spanish sparrows in Europe and Northern Africa.

Italian sparrow House

sparrow Spanish

sparrow

Aims

The over-arching aim of my thesis has been to gain a better understanding of the hybrid speciation process at a genomic level. When I began my work, we had just scratched the surface of the Italian sparrow hybrid system. It had been shown that the Italian sparrow is phenotypically and genetically mosaic (Summers-Smith 1988; Hermansen et al. 2011; Elgvin et al. 2011), yet little was known about the nature of reproductive barriers between the hybrid and its parents. While Hermansen et al. 2011 suggested there was evidence for partial reproductive isolation, this was inferred from the allele frequencies of micro-satellite markers with limited power in di!erentiating species.

Therefore, the "rst goal of my PhD research in Paper I was to test whether or not there is evidence for reproductive barriers between the Italian sparrow and its parents in zones of contact. This would serve to both solidify the Italian sparrow’s status as a hybrid species and provide insight into how the Italian sparrow has maintained as its own distinct, hybrid lineage. Once we con"rmed that there was evidence for reproductive isolation between the Italian sparrow and its parents, the next step was to investigate the hybrid system on a larger scale. Understanding which genomic regions were inherited from either parent, where there is evidence of selection and identifying areas of the genome potentially involved in reproductive isolation would provide further insight into how the Italian sparrow became its own species. Thus, in Paper II, we sought to characterize the Italian sparrow in relation to its parents by comparing entire genomes of the three focal taxa.

Finally, in Paper III, the goal was to extend our genomic

understanding of the hybrid system by comparing multiple geographically

isolated populations of the Italian sparrow on di!erent Mediterranean

islands. Investigating which genomic regions were invariably inherited from

one parent or the other, would help in identifying areas that may be

constrained for the formation of a functional hybrid, as well as which regions

are more free to vary. Together, these papers shed light onto the hybrid

speciation process by providing a detailed investigation into the genome of Italian sparrow and highlighting regions that may have been important in the formation of species barriers that have enabled the Italian sparrow to remain distinct from its parents.

Paper Summaries

Paper I – Evidence for mito-nuclear and sex-linked reproductive barriers between the hybrid Italian sparrow and its parent species

In this paper, we tested if there was evidence of reproductive barriers between the Italian sparrow and its parent species at geographic range boundaries. We "rst utilized whole-transcriptome sequencing of the house and Spanish sparrows to identify 86 parent species diagnostic SNP markers.

We then sequenced these markers in Italian sparrows (n=385) from

populations across its range throughout Italy and Sicily including contact

zones with both parents at its range boundaries. Spanish sparrows (n=142)

from Spain, Sardinia and a Spanish/Italian sympatric zone in southeast Italy,

and house sparrows (n=85) from the Czech Republic and Norway were also

sequenced for the same SNP set. We employed Bayesian genomic and

geographic cline analyses to identify markers exhibiting steep clines at

range boundaries, indicative of a decrease in introgression of a parent’s

alleles into the Italian sparrow’s genetic background. We demonstrated that

there is evidence for post-zygotic reproductive barriers between the hybrid

and its parents and identi"ed seven markers that exhibited the steepest

clines as candidate loci involved in reproductive isolation. A

disproportionately large number of these candidate loci were found on the Z

chromosome. Also, the mitochondria and nuclear genes with mitochondrial function demonstrated patterns of reduced introgression of Spanish sparrow alleles in the Italian sparrow. We conclude mito-nuclear incompatibilities isolate Italian and Spanish sparrows while sex-linked incompatibilities isolate Italian and house sparrows. We also found no evidence for hybridization between sympatric Italian and Spanish sparrows in the Gargano peninsula indicating pre-zygotic barriers may also exist between the taxa. We suggest habitat dependent assortative mating plays a part as the Italian sparrow more closely resembles the human commensal house sparrow rather than the Spanish sparrow, which occupies more rural habitats. Overall, we conclude that the mechanisms of reproductive isolation in hybrid speciation may be similar to those of non-hybrid speciation with the exception that they are against two parent species rather than one.

Paper II – The genomic mosaicism of hybrid speciation

In Paper II, the goal was to examine hybrid speciation at a whole-genome level by characterizing patterns of parental inheritance and looking for signatures of selection in the Italian sparrow genome. To accomplish this, we

"rst whole-genome sequenced and de novo assembled a reference house

sparrow genome. We then mapped whole genome data from populations of

the three focal taxa to the reference genome. Through population genetic

and admixture analyses as well as phylogenetic inference we characterized

the composition of the Italian sparrow’s genome in relation to its parents. We

found balanced yet heterogenous levels of parental contribution in the

hybrid genome and identi"ed regions where the Italian sparrow exhibits

divergence from both parent species. Areas of novel divergence in the Italian

sparrow lineage demonstrated patterns of variation consistent with

balancing selection suggesting a heterozygotic advantage in the hybrid and

were enriched for genes involved in immune system regulation. We

speculate that the admixed Italian sparrow genome may have had an advantageous e!ect on "tness thereby facilitating its spread. Furthermore, we identi"ed regions of high divergence between the hybrid and each of its parents. These regions were disproportionately located on the Z chromosome and overrepresented in gene networks likely to be involved in reproductive barriers and species-speci"c adaptations such as body patterning and beak morphology. Additionally, we found evidence for heteroplasmy with mitochondrial sequences from both house and Spanish sparrows in two Italian sparrows. We conclude that both a mosaic pattern of parental inheritance and novel divergence in the hybrid lineage have contributed to the formation of the Italian sparrow.

Paper III – Variation and constraints in hybrid genome formation

In this paper, we investigated the extent to which the Italian sparrow genome varies between populations and if there is evidence for constraints in hybrid genome formation. We used whole-genome data from four Mediterranean island populations (Crete, Corsica, Sicily and Malta) of Italian sparrows with divergent morphologies to test this by comparing patterns of admixture and di!erentiation between them. We show that there is variation in the genomic combinations making up the Italian sparrow genome and suggest that these novel hybrid genomic combinations may have arisen independently more than once in the Mediterranean islands. In particular, we "nd that the islands di!erentiate in regions holding candidate genes for beak shape and plumage color and suggest that the di!ering combinations of parental genomes allowed for adaptive di!erentiation between the isolated island populations.

We also "nd that some areas of the genome are inherited invariably from one

of the parent species indicating the presence of genomic constraints. Mito-

nuclear genes and genes involved in DNA repair were strongly constrained

to house sparrow inheritance indicating the importance of incompatibilities

in reproductive isolation between Spanish and Italian sparrows. There were fewer genes invariably inherited from the Spanish sparrow, but the ones that were identi"ed a!ected external phenotype rather than genome function. In general, these constrained regions are over-represented on the Z chromosome, consistent with the pattern of reduced introgression on sex chromosomes. This paper demonstrates that while there are varying genomic combinations that can lead to a functional hybrid genome which likely facilitate local adaptation, some regions are constrained in parallel producing strong reproductive barriers against one parent.

Discussion

Patterns of admixture in a hybrid species

To date, the Italian sparrow represents one of the few homoploid hybrid species studied in depth with genomic data. The papers in my dissertation help shed light on the processes at play in this poorly understood mode of speciation. First and foremost, my work helps to further validate the Italian sparrow’s status as a hybrid species by demonstrating evidence of reproductive isolation as well as extensive, genome-wide admixture from its parents. The Italian sparrow’s mosaic pattern of parental inheritance more closely resembles what has been observed in Helianthus sun#owers (Rieseberg et al. 2003) and the tiger swallow tail butter#ies (Kunte et al. 2011) than hybrid species shown to di!er in only small genomic regions such as the Heliconius butter#ies (Heliconius Genome Consortium 2012). Patterns of mosaicism across a hybrid’s genome are likely a product of both the age of the hybrid species and degree of divergence between the parental lineages.

The Italian sparrow is a hybrid species proposed to have arisen

around 8,000 years ago when the house sparrow's range rapidly expanded

throughout Europe during the spread of agriculture with the Neolithic

revolution (Hermansen et al. 2011; Sætre et al. 2012). The parental house and Spanish sparrows are inferred to have split from a common ancestor < 1 million years ago (Ravinet et al. 2018). At later stages of speciation, high levels of divergence are expected to be maintained genome-wide while introgression is localized to small genomic regions (Ravinet et al. 2018).

Therefore, the largely heterogenous landscape of parental inheritance in the Italian sparrow may re#ect the fact that the parent species themselves are not highly divergent. Larger block-like patterns of inheritance are observed on the Z chromosome in the Italian sparrow, which is consistent with recombination rate variation and selection shaping patterns of hybrid ancestry. In areas of low recombination, BDM incompatibilities are expected to be purged along with larger surrounding areas of neutral variation, producing block-like inheritance patterns. In contrast, in areas of high recombination or weak selection, haplotype blocks are expected to be broken apart more easily, allowing for decoupling of neutral variation from incompatibility loci (Schumer et al. 2018). Furthermore, the Z chromosome also has a lower recombination rate since recombination occurs only in males. The larger ancestry blocks on the Z chromosome relative to autosomes in the Italian sparrow supports previous studies (Carling and Brum"eld 2008; Ellegren et al. 2012) in "nding that sex chromosomes are at a later stage of speciation than the rest of the genome.

While the Italian sparrow maintains large portions of its genome from

both parents, the amount from each respective parent is variable re#ecting

the fact that there is a spectrum of genomic parental contributions within the

hybrid lineage as demonstrated in Paper III. The relatively low divergence

between the parents may mean that few incompatibilities exist between

them and occur in their hybrid o!spring. With fewer incompatibilities in need

of purging from the hybrid lineage, multiple functional hybrid genomic

combinations are possible. The papers in this PhD provide an in-depth look

at admixture across the genome of a hybrid species helping to elucidate how

hybridization can shape genomes.

Reproductive isolation from the parents

Escape from genetic incompatibilities is believed to be easier when the parent species are more closely related (Orr et al. 1997; Tubaro and Lijtmaer 2002), yet this may in turn make the establishment of reproductive isolation between the hybrid and its parents more di$cult. The Italian sparrow may represent a good balance of these conditions as Paper III demonstrates that there are variable genomic combinations that constitute the Italian sparrow, while some areas are genomically constrained to inheritance from one parent. In Paper I, we found evidence for reproductive isolation between the Italian sparrow in zones of contact and identi"ed candidate reproductive isolation loci. Subsequently, (Hermansen et al. 2014) found that these candidate reproductive isolation loci are a subset of loci isolating the parent species. This suggests the Italian sparrow has sorted incompatibilities isolating its parents in order to form reproductive barriers against both.

In line with many previous studies (Macholán et al. 2007; Carling and Brum"eld 2008; Teeter et al. 2010; Gompert and Buerkle 2011b; Ellegren et al. 2012; Taylor et al. 2014), the papers in this dissertation "nd candidate incompatibility loci were disproportionately found on the Z chromosome.

The Z chromosome has a reduced e!ective population size due to female heterogamety, which is expected to increase the rate of "xations via drift or selection. However, selection tests on coding sequences and Tajima’s D estimates are consistent with a role for selection on the Z chromosome.

Together, the papers in this thesis provide support for sex chromosomes being paramount in establishing reproductive barriers during hybrid speciation.

In Papers I and III, we also "nd evidence for less well-documented mito-nuclear incompatibilities playing a crucial part in isolating the Spanish and Italian sparrow. There is strong selection for compatibility between the nuclear and mitochondrial genomes (Hill 2016) as mito-nuclear dysfunction can have severe consequences on "tness (Camus et al. 2015; Patel et al.

2016). In fact, other hybridizing bird species have been shown to experience

reduced introgression of mitochondrial loci compared to nuclear loci (Kvist and Rytkoenen 2006; Carling and Brum"eld 2008; Taylor et al. 2013).

Moreover, "tness costs associated with suboptimal respiration have been found in hybridizing bird species (Olson et al. 2010; McFarlane et al. 2016).

The majority of Italian sparrows have inherited solely house sparrow mtDNA and I found evidence that there is also strong selection against Spanish sparrow inheritance in nuclear genes with mitochondrial function. In Paper I, we suggest this pattern is consistent with patterns of genomic con#ict in the form of ‘mothers curse’. Since mitochondria are maternally inherited, male detrimental mutations will not be subject to selection and female advantageous, yet male detrimental mutations can accumulate (Beekman et al. 2014). This e!ect can be remedied in males through compensatory mutations which is expected to arise on the Z chromosome where male- speci"c "tness e!ects are more likely (Connallon et al. 2018). This is consistent with our "ndings that candidate reproductive isolation loci were predominantly sex-linked. Paternal leakage is another suggested mechanism for overcoming negative "tness e!ects of maternally inherited mitochondria.

Notably, heteroplasmy has also been observed in hybridizing taxa (Shitara et al. 1998; Kvist et al. 2003; Radojicic et al. 2015; Śmietanka and Burzyński 2017), which we found evidence for in Papers II and III.

While mito-nuclear incompatibilities were shown to be instrumental in isolating the Italian and Spanish sparrows, Paper III found that genes invariably inherited from the Spanish sparrow as candidates for a!ecting external phenotype and included a candidate gene for plumage color. A role for pre-zygotic barriers in isolating the Italian sparrow from the house sparrow may therefore be important. This has been supported by a cline analysis in the Alps hybrid zone between house and Italian sparrows demonstrating selection on plumage traits (Bailey et al. 2015).

In all, my work provides evidence for the existence of reproductive

barriers between the Italian sparrow and its parents and highlights the

importance of sex-linked and mito-nuclear incompatibilities in the formation

of hybrid species.

Selection within the hybrid lineage

The papers in this dissertation also provide evidence of selection within the hybrid lineage. One proposed mechanism of hybrid speciation is that a hybrid maintains "tness advantages compared to the parents by being able to occupy a new ecological niche (Mallet 2007) and that hybrids may even displace their parents if they experience higher "tness in a given habitat (Buerkle et al. 2000). Both Papers II and III "nd evidence for balancing selection in areas where the Italian sparrow is divergent from both parent taxa and in Paper II these regions were enriched for genes with functions in the immune system. We postulate that the admixed genome may also have an advantageous e!ect in traits with an intermediate optimum and that this may have facilitated its spread. Genomic regions where the morphologically divergent populations of Italian sparrows di!erentiated from each other were also found to harbor candidate loci for beak shape, feather development and melanogenesis underscoring the fact that novel variation combined into admixed genomes can allow for adaptive di!erentiation between isolated populations of hybrid species.

Future perspectives & concluding remarks

Though my work provides a glimpse of how hybrid speciation can occur at a genomic level, there are still many avenues of further research to help elucidate the hybrid speciation process. One question that remains to be answered is how the timing and magnitude of hybridization events a!ect speciation potential. Phylogenetic analysis of mitochondrial markers in the system have supported the hypothesis that the Italian sparrow arose as a result of human activity during the Neolithic revolution (Hermansen et al.

2011; Sætre et al. 2012), however, further research is needed to more

robustly test this. More speci"cally, model based estimation of the timing

and magnitude of gene #ow would provide further insight into the hybrid speciation process (Payseur and Rieseberg 2016).

Additionally, while my work points to candidate regions and genes involved in reproductive isolation, the functional genetics behind these proposed areas is still unknown. Further research into the phenotypic e!ects of these regions, and the identi"cation of the speci"c genes and causative factors is needed. It would also be interesting to investigate if structural variation plays a role in reproductive isolation in the system.

Inversions have been shown to rapidly drive species divergence (Noor et al.

2001) and combining two parental genomes into a hybrid genome may result in structural variation potentially forming powerful reproductive barriers. In fact, both pre- and post-zygotic reproductive barriers have been shown to map to inversions in species pairs (Kirkpatrick 2010).

Another question raised from my work and an avenue for further research is the role and extent of heteroplasmy in Italian sparrows. If mito- nuclear incompatibilities are instrumental as reproductive barriers, then how does heteroplasmic mitochondria "t into the picture? The system would bene"t from a more detailed investigation into how heteroplasmic individuals di!er from the Italian sparrows with solely house sparrow mitochondria in terms of "tness, patterns of parental inheritance in nuclear-encoded mitochondrial genes and degree of isolation from the parents.

Finally, another line of research that needs to be further explored is the role of ecology and geography in shaping hybrid traits in the Italian sparrow. In Paper II, we suggest that the hybrid genome may have experienced "tness advantages facilitating its spread, one of the proposed mechanisms for hybrid success (Buerkle et al. 2000; Mallet 2007).

Additionally, Paper III demonstrated there is evidence for adaptive

di!erentiation among island populations of Italian sparrows and beak shape

has been shown to be strongly in#uenced by precipitation regimes in Italy

(Eroukhmano! et al. 2013). This points to a role of ecology and local

adaptation in the formation of the Italian sparrow. However, the

phylogeographic history of a species is important in understanding how new

species can successfully spread, and combining it with phenotypic and

genomic data is needed to understand how species are able to occupy new niches (Trucchi et al. 2016). Thus, investigating the phylogeographic history of the Italian sparrow and pairing it with the morphological, ecological and genomic data would be powerful way to move forward and investigate how the Italian sparrow has been successful in occupying the Italian peninsula and surrounding islands.

Overall, throughout my PhD research, I have used genomic data to paint a picture as to how hybridization between two lineages has the potential to create a novel species. My work serves as a detailed investigation of an example of homoploid hybrid speciation in animals and provides a jumping o! point for future hybrid speciation research. It is an exciting time as fast developing technology has enabled new approaches for studying hybrid speciation and as more information becomes available, it is becoming clear that hybridization is a powerful force of variation and can be a catalyst for speciation.

Acknowledgements

Many thanks to Tore Elgvin, Fabrice Eroukhmano! and Mark Ravinet for taking the time to read my thesis and provide helpful comments. Also a big thank you to Anna Mazzarella for advice with layout editing.

References

Abbott, R. J., N. H. Barton, and J. M. Good. 2016. Genomics of hybridization and its evolutionary consequences. Molecular Ecology 25:2325–2332.

Abbott, R.J., D. Albach, S. Ansell, J. W. Arntzen, S. J. E. Baird, N. Bierne, J.

Boughman, A. Brelsford, C. A. Buerkle, R. Buggs, R. K. Butlin, U.

Dieckmann, F. Eroukhmano!, A. Grill, S. H. Cahan, J. S. Hermansen, G.

Hewitt, A. G. Hudson, C. Jiggins, J. Jones, B. Keller, T. Marczewski, J.

Mallet, P. Martinez-Rodriguez, M. Möst, S. Mullen, R. Nichols, A. W.

Nolte, C. Parisod, K. Pfennig, A. M. Rice, M. G. Ritchie, B. Seifert, C. M.

Smadja, R. Stelkens, J. M. Szymura, R. Vainola, J. B. W. Wolf, and D.

Zinner. 2013. Hybridization and speciation. Journal of Evolutionary Biology 26:229–246.

Arnegard, M. E., M. D. McGee, B. Matthews, K. B. Marchinko, G. L. Conte, S.

Kabir, N. Bedford, S. Bergek, Y. F. Chan, F. C. Jones, D. M. Kingsley, C. L.

Peichel, and D. Schluter. 2014. Genetics of ecological divergence during speciation. Nature 511:307–311.

Arnold, M. L., and S. A. Hodges. 1995. Are natural hybrids "t or un"t relative to their parents? Trends in Ecology and Evolution 10:67–71.

Baack, E. J., and L. H. Rieseberg. 2007. A genomic view of introgression and hybrid speciation. Current Opinion in Genetics and Development 17:513–518.

Bailey, R. I., M. R. Tesaker, C. N. Trier, and G. P. Saetre. 2015. Strong selection on male plumage in a hybrid zone between a hybrid bird species and one of its parents. Journal of Evolutionary Biology 28:1257–1269.

Ballard, J., and D. M. Rand. 2005. The population biology of mitochondrial DNA and its phylogenetic implications. Annual Review of Ecology Evolution and Systematics 36:621–642.

Bar-Yaacov, D., Z. Hadjivasiliou, L. Levin, G. Barshad, R. Zarivach, A. Bouskila, and D. Mishmar. 2015. Mitochondrial involvement in vertebrate

speciation? The Case of mito-nuclear genetic divergence in chameleons. Genome Biology and Evolution 7:3322–3336.

Barton, N. H. 2001. The role of hybridization in evolution. Molecular Ecology 10:551–568.

Barton, N. H., and G. M. Hewitt. 1989. Adaptation, speciation and hybrid zones. Nature 341:497–503.

Barton, N. H., and G. M. Hewitt. 1985. Analysis of hybrid zones. Annual Review of Ecology and Systematics 16:113–148.

Bateson, W. 1909. Heredity and variation in modern lights. Darwin and Modern Science: 85-101.

Beekman, M., D. K. Dowling, and D. K. Aanen. 2014. The costs of being male:

are there sex-speci"c e!ects of uniparental mitochondrial inheritance? Philosophical Transactions of the Royal Society of London B: Biological Sciences 369:1–7.

Buerkle, C. A., and C. Lexer. 2008. Admixture as the basis for genetic mapping. Trends in Ecology and Evolution 23:686–694.

Buerkle, C. A., and L. H. Rieseberg. 2001. Low intraspeci"c variation for genomic isolation between hybridizing sun#ower species. Evolution 55:684–691.

Buerkle, C. A., and L. H. Rieseberg. 2008. The rate of genome stabilization in homoploid hybrid species. Evolution 62:266–275.

Buerkle, C. A., R. J. Morris, M. A. Asmussen, and L. H. Rieseberg. 2000. The likelihood of homoploid hybrid speciation. Heredity 84:441–451.

Bullini, L. 1994. Origin and evolution of animal hybrid species. Trends in Ecology and Evolution 9:422–426.

Burri, R., A. Nater, T. Kawakami, C. F. Mugal, P. I. Olason, L. Smeds, A. Suh, L.

Dutoit, S. Bureš, L. Z. Garamszegi, S. Hogner, J. Moreno, A.

Qvarnström, M. Ružić, S.-A. Sæther, G.-P. Sætre, J. Toeroek, and H.

Ellegren. 2015. Linked selection and recombination rate variation drive the evolution of the genomic landscape of di!erentiation across the speciation continuum of Ficedula #ycatchers. Genome Research 25:1656–1665.

Camus, M. F., J. B. W. Wolf, E. H. Morrow, and D. K. Dowling. 2015. Single

nucleotides in the mtDNA sequence modify mitochondrial molecular function and are Associated with sex-speci"c e!ects on fertility and aging. Current Biology 25:2717–2722.

Carling, M. D., and R. T. Brum"eld. 2008. Haldane's rule in an avian system:

Using cline theory and divergence population genetics to test for di!erential introgression of mitochondrial, autosomal, and sex-linked loci across the passerinabunting hybrid zone. Evolution 62:2600–

2615.

Carneiro, M., S. J. E. Baird, S. Afonso, E. Ramirez, P. Tarroso, H. Teotónio, R.

Villafuerte, M. W. Nachman, and N. Ferrand. 2013. Steep clines within a highly permeable genome across a hybrid zone between two

subspecies of the European rabbit. Molecular Ecology 22:2511–2525.

Charlesworth, B., J. A. Coyne, and N. H. Barton. 1987. The relative rates of evolution of sex chromosomes and autosomes. The American Naturalist 130:113–146.

Connallon, T., M. F. Camus, E. H. Morrow, and D. K. Dowling. 2018.

Coadaptation of mitochondrial and nuclear genes, and the cost of mother's curse. Proceedings of Biological Science. 285:e20172257.

Coyne, J. A., and H.A. Orr. 2004. Speciation. Sinauer Associates, Inc., Sunderland (MA).

Dobzhansky, T. 1948. Genetics of Natural Populations. Genetics 33:588–602.

Dobzhansky, T. 1936. Studies on hybrid sterility. II. Localization of sterility factors in Drosophila pseudoobscura hybrids. Genetics 21:113–135.

Dowling, T. E., and A. C. L. Secor. 1997. The role of hybridization and introgression in the diversi"cation of animals. Annual Review of Ecology and Systematics 28:593–619.

Elgvin, T. O., J. S. Hermansen, A. Fijarczyk, T. Bonnet, T. Borge, S. A. Sæther, K.

L. Voje, and G.-P. Sætre. 2011. Hybrid speciation in sparrows II: a role for sex chromosomes? Molecular Ecology 20:3823–3837.

Ellegren, H., L. Smeds, R. Burri, P. I. Olason, N. Backström, T. Kawakami, A.

Künstner, H. Mäkinen, K. Nadachowska-Brzyska, A. Qvarnström, S.

Uebbing, and J. B. W. Wolf. 2012. The genomic landscape of species divergence in Ficedula #ycatchers. Nature 491:756–760.

Eroukhmano!, F., J. S. Hermansen, R. I. Bailey, S. A. Sæther, and G.-P. Sætre.

2013. Local adaptation within a hybrid species. Heredity 111:286–292.

Feliner, G.N., M. Álvarez, J. Fuertes-Aguilar, M. Heuertz, I. Marques,

F. Moharrek, R Piñeiro, R Riina, J.A.Rosselló, P.S.Soltis, I. Villa-

Machío 2017. Is homoploid hybrid speciation that rare? An

empiricist’s view. Heredity 118: 513-516.

Felsenstein, J. 1981. Skepticism towards Santa Rosalia, or why are there so few kinds of animals? Evolution 35:124–138.

Fisher, R.A., 1930. The genetical theory of natural selection. Clarendon Press, Oxford.

Gompert, Z., and C. A. Buerkle. 2011a. A hierarchical bayesian model for next-generation population genomics. Genetics 187:903–917.

Gompert, Z., and C. A. Buerkle. 2011b. Bayesian estimation of genomic clines. Molecular Ecology 20:2111–2127.

Gompert, Z., L. K. Lucas, C. C. Nice, J. A. Fordyce, M. L. Forister, and C. A.

Buerkle. 2012. Genomic regions with a history of divergent selection a!ect "tness of hybrids between two butter#y species. Evolution 66:2167–2181.

Good, J. M., D. Vanderpool, S. Keeble, and K. Bi. 2015. Negligible nuclear introgression despite complete mitochondrial capture between two species of chipmunks. Evolution 69:1961–1972.

Grant, P. R., and B. R. Grant. 1994. Phenotypic and genetic e!ects of hybridization in Darwin's "nches. Evolution 48:297.

Grant, P. R., B. R. Grant, and K. Petren. 2015. Hybridization in the recent past.

The American Naturalist 166:56–67.

Gross, B. L., and L. H. Rieseberg. 2005. The ecological genetics of homoploid hybrid speciation. Journal of Heredity 96:241–252.

Harrison, R. G. 1993. Hybrid zones and the evolutionary process. Oxford University Press, New York, NY.

Harrison, R. G., and E. L. Larson. 2014. Hybridization, introgression, and the nature of species boundaries. Journal of Heredity 105:795–809.

Hegarty, M. J., and S. J. Hiscock. 2008. Genomic clues to the evolutionary success of review polyploid plants. Current Biology 18:R435–R444.

Heliconius Genome Consortium. 2012. Butter#y genome reveals promiscuous exchange of mimicry adaptations among species.

Nature 487:94–98.

Hermansen, J. S., F. Haas, C. N. Trier, R. I. Bailey, A. J. Nederbragt, A. Marzal, and G.-P. Sætre. 2014. Hybrid speciation through sorting of parental incompatibilities in Italian sparrows. Molecular Ecology 23:5831–5842.

Hermansen, J. S., S. A. Saether, T. O. Elgvin, T. Borge, E. Hjelle, and G.-P.

Sætre. 2011. Hybrid speciation in sparrows I: phenotypic

intermediacy, genetic admixture and barriers to gene #ow. Molecular Ecology 20:3812–3822.

Hill, G. E. 2016. Mitonuclear coevolution as the genesis of speciation and the mitochondrial DNA barcode gap. Ecology and Evolution 6:5831–5842.

Hill, G. E. 2017. The mitonuclear compatibility species concept. The Auk 134:393–409.

Jiggins, C. D., C. Salazar, M. Linares, and J. Mavarez. 2008. Hybrid trait speciation and Heliconius butter#ies. Philosophical Transactions of the Royal Society of London B: Biological Sciences 363:3047–3054.

Keller, I., C. E. Wagner, L. Greuter, S. Mwaiko, O. M. Selz, A. Sivasundar, S.

Wittwer, and O. Seehausen. 2013. Population genomic signatures of

divergent adaptation, gene #ow and hybrid speciation in the rapid radiation of Lake Victoria cichlid "shes. Molecular Ecology 22:2848–

2863.

Kirkpatrick, M. 2010. How and why chromosome inversions evolve. PLoS Biology 8:e1000501

Kunte, K., C. Shea, M. L. Aardema, J. M. Scriber, T. E. Juenger, L. E. Gilbert, and M. R. Kronforst. 2011. Sex chromosome mosaicism and hybrid speciation among tiger swallowtail butter#ies. PLoS Genetics 7:e1002274–14.

Kvist, L., and S. Rytkoenen. 2006. Characterization of a secondary contact zone of the great tit Parus major and the Japanese tit p-minor (Aves : Passeriformes) in far eastern Siberia with DNA markers. Zootaxa 55–

Kvist, L., J. Martens, A. A. Nazarenko, and M. Orell. 2003. Paternal 73.

leakage of mitochondrial DNA in the great tit (Parus major). Molecular Biology and Evolution. 20:243–247.

Lamichhaney, S., F. Han, M. T. Webster, L. Andersson, B. R. Grant, and P. R.

Grant. 2018. Rapid hybrid speciation in Darwin’s "nches. Science 359:224–228.

Larsen, P. A., M. R. Marchan-Rivadeneira, and R. J. Baker. 2010. Natural hybridization generates mammalian lineage with species

characteristics. Proceedings of the National Academy of Science U.S.A. 107:11447–11452.

Macholán, M., P. Munclinger, M. Šugerková, P. Dufková, B. Bímová, E.

Božíková, J. Zima, and J. Piálek. 2007. Genetic analysis of autosomal and X-linked markers across a mouse hybrid zone. Evolution 61:746–

771.

Mallet, J. 1995. A species de"nition for the modern synthesis. Trends in Ecology and Evolution 10:294–299.

Mallet, J. 2007. Hybrid speciation. Nature 446:279–283.

Mallet, J. 2005. Hybridization as an invasion of the genome. Trends in Ecology and Evolution 20:229–237.

Mank, J. E., B. Vicoso, S. Berlin, and B. Charlesworth. 2010. E!ective population size and the faster-X e!ect: Empirical results and their interpretation. Evolution 64:663–674.

Mavarez, J., and M. Gonzalez. 2006. A set of microsatellite markers for

Heliconius melpomene and closely related species. Molecular Ecology Notes 6:20–23.

Mavarez, J., and M. Linares. 2008. Homoploid hybrid speciation in animals.

Molecular Ecology 17:4181–4185.

Mavarez, J., C. A. Salazar, E. Bermingham, C. Salcedo, C. D. Jiggins, and M.

Linares. 2006. Speciation by hybridization in Heliconius butter#ies.

Nature 441:868–871.

May, R. M., J. A. Endler, and R. E. McMurtrie. 2015. Gene frequency clines in the presence of selection opposed by gene #ow. The American

Naturalist 109:659–676.